INDEX

Explanations

references to specific locations and regional entities

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

US and German locations

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

OnClickListener

-0.52

weise

-0.50

multirow

-0.49

mat

-0.48

ToInt

-0.47

meg

-0.46

Piper

-0.46

蒲

-0.45

 卡片

-0.45

뢰

-0.45

POSITIVE LOGITS

Particular

0.85

 يتيمه

0.77

PropertyChanging

0.74

initComponents

0.74

UB

0.73

 estekak

0.73

writeFieldEnd

0.72

érales

0.72

 handleMessage

0.71

neous

0.71

Activations Density 0.200%

references to specific locations and regional entities

US and German locations

No Known Activations